10 May, 2020

h2.title { font-size: 8px; #color: #a9a9a9; text-align: center; }

## Warning: package 'tibble' was built under R version 3.6.2

Introduction

Dataset:

  • breast cancer
  • proteomics by mass spectrometry

Goal:

  • Explore the dataset for patterns

  • Create models to identify the breast cancer subclasses

Materials and Methods

Dataset:

Materials and Methods:

*Exploratory data analysis of clinical data

  • PCA

  • K-means

  • ANN

No definitive effects between expression landscapes and specific tumor subclasses

Breast cancer subtypes in the dataset are well represented

Breast cancer subtypes do not discriminate

Breast cancer & Gender

Dimentionality reduction

PCA analysis

K-means clustering

ANN model's representation

ANN representation

Current "file" state

Discussion

  • What could have been better

  • further work

End 2

End 3

End 1